On Saturday, September 17, 2022 at 09:11:26 p.m. EDT, gene heskett
<[email protected]> wrote:
Greetings, got a new idea;
I am currently doing half the "make it round" part of my screw code
without any taper comp. to
correct for the axis z height diff from the driven end of the stick to
the live center. Printing
shims to put under the live center is limited in how close I can get
that because of the printers
z step height. So the best I can do there is to get it within that step
height, measure it at
both ends of the stick, which is diameter, and cut that in half cuz
we're fixing it with a radius
change.
But that tiny a tool dulls and the flexibility of the x rods etc in the
gantry means I can see the
tool rising as the corners of the square stick go by. So far I've been
fudging by making the
correction larger, but as it approaches true round there is less and
less motion because its
always being pushed away by the dulling tool.
So I got this wild idea of sampling a 5kg strain gauge to measure this
deflection, and run an
add of its reported error to push the tool back down, making the tool
follow the programmed
path a little closer.
But to do that right, at least in my mind, I've got to sample the gauge
as the peak of the original
square goes by at 4x per full turn. Or maybe sample the peak regardless
of its position.
And rig a running average in my samples. Once per rev is easy enough,
just use the B home signal
as a trigger, but how would I rig a pll to do it 4x per turn?
And make it immune to feed rate changes... A timer started by the home
switch would determine the
first sample point and an led banged by a cap shining on the stick would
help to time the first
sample of each turn, but its been decades since the last time I made a
pll do as it was told. And
I don't see a pll module in the hal menu.
Ideas? Like how to make it keep a running average over 4 samples.
I did do that on the go704, trying to make my low rez optical spindle
encoder less noisy, but the
real cure was an encoder on the motor, and tally switches on the gear
shift knob to change the Scale...
Take care & stay well everybody.
Cheers, Gene Heskett.
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
- Louis D. Brandeis
Genes Web page <http://geneslinuxbox.net:6309/>
Hello Gene.
I would monitoring the state of the four corners as they turn.On my lathe, the
spindle hole diameter, through the headstock of the lathe, is one and one eight
of an inch.This pipe (or tube) through the headstock is driven by the lathe,
and then the mounting point for what I am imagining is the the lathe dog
driving the four sided wood.I would then mount a one part in 10 to the 14
rotational (chips + electronics + magnet) on the spindle hole pipe, to tell me
where the wood is.
Then comes the "where is the point I want to measure" and "how much distance do
I put into the toolbit" programming work.If you want a kit of the stuff, I have
two of them.
If someone else has already provided that idea, I apologize.This is the first
time in a couple months I have got back to reading the emails.
James Isaac.
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users
